aaazKI commited on
Commit
138da4f
·
verified ·
1 Parent(s): cfdbe89

Upload config.json

Browse files
Files changed (1) hide show
  1. config.json +179 -0
config.json ADDED
@@ -0,0 +1,179 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "RagRetriever"
4
+ ],
5
+ "dataset": "wiki_dpr",
6
+ "dataset_split": "train",
7
+ "do_deduplication": true,
8
+ "do_marginalize": false,
9
+ "doc_sep": " // ",
10
+ "exclude_bos_score": false,
11
+ "generator": {
12
+ "_num_labels": 3,
13
+ "activation_dropout": 0.0,
14
+ "activation_function": "gelu",
15
+ "add_bias_logits": false,
16
+ "add_cross_attention": false,
17
+ "add_final_layer_norm": false,
18
+ "architectures": [
19
+ "BartModel",
20
+ "BartForMaskedLM",
21
+ "BartForSequenceClassification"
22
+ ],
23
+ "attention_dropout": 0.0,
24
+ "bad_words_ids": null,
25
+ "bos_token_id": 0,
26
+ "chunk_size_feed_forward": 0,
27
+ "classif_dropout": 0.0,
28
+ "d_model": 1024,
29
+ "decoder_attention_heads": 16,
30
+ "decoder_ffn_dim": 4096,
31
+ "decoder_layerdrop": 0.0,
32
+ "decoder_layers": 12,
33
+ "decoder_start_token_id": 2,
34
+ "do_sample": false,
35
+ "dropout": 0.1,
36
+ "early_stopping": false,
37
+ "encoder_attention_heads": 16,
38
+ "encoder_ffn_dim": 4096,
39
+ "encoder_layerdrop": 0.0,
40
+ "encoder_layers": 12,
41
+ "eos_token_id": 2,
42
+ "extra_pos_embeddings": 2,
43
+ "finetuning_task": null,
44
+ "force_bos_token_to_be_generated": false,
45
+ "id2label": {
46
+ "0": "LABEL_0",
47
+ "1": "LABEL_1",
48
+ "2": "LABEL_2"
49
+ },
50
+ "init_std": 0.02,
51
+ "is_decoder": false,
52
+ "is_encoder_decoder": true,
53
+ "label2id": {
54
+ "LABEL_0": 0,
55
+ "LABEL_1": 1,
56
+ "LABEL_2": 2
57
+ },
58
+ "length_penalty": 1.0,
59
+ "max_length": 20,
60
+ "max_position_embeddings": 1024,
61
+ "min_length": 0,
62
+ "model_type": "bart",
63
+ "no_repeat_ngram_size": 0,
64
+ "normalize_before": false,
65
+ "normalize_embedding": true,
66
+ "num_beams": 1,
67
+ "num_hidden_layers": 12,
68
+ "num_return_sequences": 1,
69
+ "output_attentions": false,
70
+ "output_hidden_states": false,
71
+ "output_past": false,
72
+ "pad_token_id": 1,
73
+ "prefix": " ",
74
+ "pruned_heads": {},
75
+ "repetition_penalty": 1.0,
76
+ "return_dict": false,
77
+ "scale_embedding": false,
78
+ "static_position_embeddings": false,
79
+ "task_specific_params": {
80
+ "summarization": {
81
+ "early_stopping": true,
82
+ "length_penalty": 2.0,
83
+ "max_length": 142,
84
+ "min_length": 56,
85
+ "no_repeat_ngram_size": 3,
86
+ "num_beams": 4
87
+ }
88
+ },
89
+ "temperature": 1.0,
90
+ "tie_encoder_decoder": false,
91
+ "tie_word_embeddings": true,
92
+ "tokenizer_class": null,
93
+ "top_k": 50,
94
+ "top_p": 1.0,
95
+ "torchscript": false,
96
+ "use_bfloat16": false,
97
+ "use_cache": true,
98
+ "vocab_size": 50265,
99
+ "xla_device": null
100
+ },
101
+ "index_name": "legacy",
102
+ "index_path": "zpbrent/RagReuse",
103
+ "is_encoder_decoder": true,
104
+ "label_smoothing": 0.0,
105
+ "max_combined_length": 300,
106
+ "model_type": "rag",
107
+ "n_docs": 5,
108
+ "output_retrieved": false,
109
+ "passages_path": null,
110
+ "question_encoder": {
111
+ "add_cross_attention": false,
112
+ "architectures": [
113
+ "DPRQuestionEncoder"
114
+ ],
115
+ "attention_probs_dropout_prob": 0.1,
116
+ "bad_words_ids": null,
117
+ "bos_token_id": null,
118
+ "chunk_size_feed_forward": 0,
119
+ "decoder_start_token_id": null,
120
+ "do_sample": false,
121
+ "early_stopping": false,
122
+ "eos_token_id": null,
123
+ "finetuning_task": null,
124
+ "gradient_checkpointing": false,
125
+ "hidden_act": "gelu",
126
+ "hidden_dropout_prob": 0.1,
127
+ "hidden_size": 768,
128
+ "id2label": {
129
+ "0": "LABEL_0",
130
+ "1": "LABEL_1"
131
+ },
132
+ "initializer_range": 0.02,
133
+ "intermediate_size": 3072,
134
+ "is_decoder": false,
135
+ "is_encoder_decoder": false,
136
+ "label2id": {
137
+ "LABEL_0": 0,
138
+ "LABEL_1": 1
139
+ },
140
+ "layer_norm_eps": 1e-12,
141
+ "length_penalty": 1.0,
142
+ "max_length": 20,
143
+ "max_position_embeddings": 512,
144
+ "min_length": 0,
145
+ "model_type": "dpr",
146
+ "no_repeat_ngram_size": 0,
147
+ "num_attention_heads": 12,
148
+ "num_beams": 1,
149
+ "num_hidden_layers": 12,
150
+ "num_return_sequences": 1,
151
+ "output_attentions": false,
152
+ "output_hidden_states": false,
153
+ "pad_token_id": 0,
154
+ "prefix": null,
155
+ "projection_dim": 0,
156
+ "pruned_heads": {},
157
+ "repetition_penalty": 1.0,
158
+ "return_dict": false,
159
+ "task_specific_params": null,
160
+ "temperature": 1.0,
161
+ "tie_encoder_decoder": false,
162
+ "tie_word_embeddings": true,
163
+ "tokenizer_class": null,
164
+ "top_k": 50,
165
+ "top_p": 1.0,
166
+ "torchscript": false,
167
+ "type_vocab_size": 2,
168
+ "use_bfloat16": false,
169
+ "use_cache": true,
170
+ "vocab_size": 30522,
171
+ "xla_device": null
172
+ },
173
+ "reduce_loss": false,
174
+ "retrieval_batch_size": 8,
175
+ "retrieval_vector_size": 768,
176
+ "title_sep": " / ",
177
+ "use_dummy_dataset": false,
178
+ "vocab_size": null
179
+ }